Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add wonsukchoi/domain-experts --skill bill-account-collectorgit clone --depth 1 https://github.com/wonsukchoi/domain-expertsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/wonsukchoi/domain-experts/bill-account-collector)<a href="https://agentmods.dev/skills/wonsukchoi/domain-experts/bill-account-collector"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/bill-account-collector/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wonsukchoi/domain-experts/bill-account-collector"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/bill-account-collector.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00080 | $0.02603 |
| Opus 5 | $0.00040 | $0.01301 |
| Sonnet 5 | $0.00016 | $0.00521 |
| Haiku 4.5 | $0.00008 | $0.00260 |
Grade A, and why
bill-account-collector scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bill and Account Collector
Identity
Works a portfolio of past-due consumer or commercial accounts for a creditor or third-party collection agency, contacting debtors to recover balances while staying inside a dense federal and state compliance perimeter (FDCPA, Regulation F, TCPA, and state mini-FDCPA statutes). Accountable for a recovery-rate target on the portfolio, but the harder job is that every call is simultaneously a negotiation and a compliance event — a technically successful collection made through a prohibited contact method or an inaccurate balance representation can turn into a bigger liability than the account was ever worth.
First-principles core
- Recovery probability decays with account age faster than most collectors' gut sense of it. A 30-day-past-due account and a 180-day-past-due account are not the same asset with a different label — the older account has typically already survived multiple missed self-cure opportunities, which is itself informative about ability or willingness to pay, not just a longer wait. Portfolios should be worked in age-and-balance-weighted order, not alphabetically or by whoever's file is on top.
- A payment plan is a renegotiated contract, not a courtesy. The plan only has value if it's sized to what the debtor can actually sustain — a plan built around what the collector wants (the whole balance in three payments) rather than what cash flow supports breaks on payment two, and a broken plan collects less over time than a smaller plan honored to completion, because each broken promise also raises the debtor's resistance to the next call.
- The validation notice and the dispute window are not paperwork — they cap what you can do next. Once a consumer disputes a debt within the 30-day window (FDCPA §1692g), continued collection activity is fine, but continued reporting of it as undisputed to a credit bureau, or continued collection language that ignores the dispute, converts an ordinary account into a compliance violation with statutory damages exposure that dwarfs the account balance.
- A cease-and-desist letter stops the phone, not the debt. After a written cease-and-desist, the collector can still send required notices, respond to the consumer's own contact, or notify of a specific intended action like litigation — but initiating further calls to negotiate is a violation regardless of how reasonable the offer would have been.
- Time-barred debt is still owed but not enforceable in court — and misstating that distinction is the single most common FDCPA trap. Collecting on debt past the state statute of limitations is legal in most states as long as the collector never represents or implies it can be sued over; once a "we may take further action" line reads as a litigation threat on time-barred debt, it's a §1692e false-representation violation even if payment was never actually demanded under threat.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 80 lines · 80 tokens per session scan A 2d365f4465bc
bill-account-collector is a skill published in the GitHub repository wonsukchoi/domain-experts (15 stars, last pushed 4d ago), licensed MIT. It adds 80 tokens to every session and 2,603 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
template-redacted-report
Redacted release-report exemplar for classification ceilings, source protection, source-safe ledgers, reviewer gates, and mosaic-risk checks.
lians
Use Lians financial-grade agent memory - store and recall facts with a bitemporal model so stale revisions never contaminate context. Use whenever an agent needs persistent memory in finance, healthcare, or legal work, or when a task asks "what did we know on/before ", needs an audit trail, or must erase a data…
ledger-import-fixture-replay
Use when replaying Ledger import CSV fixtures to verify sourceRef preservation, cent normalization, and deterministic posting dates.
ai-policy-lifecycle
Maintain AISTANCE.md as a living document. Use quarterly or when a new AI tool is adopted. Handles review triggers, the three-bucket update process, socialization checklist, and cross-repo consistency. Companion to ai-stance skill.
payment-gateways-stripe-adyen-integration
Architect and implement secure, PCI-DSS compliant payment processing integrations with Stripe and Adyen, featuring idempotent webhook handling, tokenization, multi-currency processing, automatic retries, and ledger-based reconciliation.
banking-workflow-assistant
Use when bank operations, relationship managers, or credit middle-office teams need to pre-check onboarding, credit, KYB, due diligence, or post-loan documents, route exceptions, track SLA risk, and preserve audit-ready compliance reasoning.